14:00:56 <licquia> #startmeeting LSB Bug Triage 2014 May 16
14:00:56 <lsbbot> Meeting started Fri May 16 14:00:56 2014 UTC.  The chair is licquia. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:56 <lsbbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:01:05 * licquia is running a little behind too
14:02:29 * mwichmann returns... not so late
14:02:57 <licquia> alrighty
14:03:03 * licquia pulls up the bug stats
14:04:56 <licquia> #info lsb: 1 reopened, 4 new
14:05:54 <licquia> #info fhs: 26 new
14:06:15 <licquia> !lsbbug 3195
14:06:18 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3195 normal, P2, ---, mats, REOPENED , LSB packages with unowned directories uninstallable on some rpm implementations.
14:06:35 <licquia> nothing to report on my end, just "we need to release that"
14:06:57 * licquia figures the updates to app-checker could go into that vip release
14:07:16 <mwichmann> sure
14:07:28 <mwichmann> if that doesn't mean these get delayed further
14:07:38 <mwichmann> "nothing to discuss here"
14:07:49 <licquia> #info no progress on 3195
14:07:59 <licquia> !lsbbug 3972
14:08:02 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3972 normal, P2, ---, licquia, NEW , Apache Tests(2.2.14-8) Fail on x86_64: Failed to prepare apache
14:08:37 * licquia checks on the latest status of the lsb init stuff
14:09:58 <mwichmann> no real changes
14:10:06 <mwichmann> file going wrong place
14:10:12 <mwichmann> if it goes right place, it's missing stuff
14:10:22 <licquia> looks like we're past that, now to problems with not starting apache
14:10:28 <mwichmann> just reported on irc yesterday that tests fail in tunrimg
14:10:29 <licquia> due to port conflict
14:10:38 <mwichmann> oh, right, forgot
14:11:13 <mwichmann> I got tired of going back and forth
14:11:16 * licquia adds a comment
14:11:35 <mwichmann> lsb-apache does use 80 (I remember stew tried an alternative but there were problems - I think I remember anyway)
14:11:57 <mwichmann> so I had tried to say - find out what else is on 80, and don't run it, but I was not clear enough I guess
14:13:00 <licquia> he thinks lsb-apache is running on port 80, so i don't know what problems he could be having
14:13:03 <licquia> so, just asked
14:13:28 <licquia> #agreed keep helping the person with appbat problems in bug 3972
14:13:54 <licquia> !lsbbug 3974
14:13:56 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3974 normal, P2, ---, licquia, NEW , LSB 4.1: Celestia Test : OpenGL GLX extension not supported by display ':0'
14:14:16 <licquia> didn't this one get solved?  missing glx extensions, or something
14:14:40 * licquia sees last comment
14:14:45 <licquia> so, close?
14:15:14 <mwichmann> wasn't paying attention
14:15:16 <mwichmann> sounds like it
14:15:25 <licquia> #agreed close 3974
14:15:48 <mwichmann> this was not-our-bug territory, we were just leaving open until they sorted
14:16:00 <licquia> yup
14:16:15 <licquia> !lsbbug 3979
14:16:17 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3979 normal, P2, ---, licquia, NEW , update distribution data
14:17:41 <licquia> so, target next vip update?  ("next" -> not necessarily the same one as above)
14:17:59 <mwichmann> I suppose
14:18:25 <mwichmann> this one is /really/ independent of LSB version
14:18:38 <licquia> yup
14:19:50 <licquia> #agreed target 3979 at next vip update (though perhaps not urgent enough to delay 3195 fix)
14:20:11 <licquia> !lsbbug 3980
14:20:14 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3980 normal, P2, ---, mats, NEW , clear typo to repair in command_tests.sh
14:20:43 <mwichmann> made changes to devel
14:21:06 <mwichmann> possibly same change to 4.1?
14:21:45 <mwichmann> can bzr cherry-pick that way, out of curiosity? (I know it's come up before)
14:21:52 <licquia> we don't actually see this bug?
14:22:14 <mwichmann> no
14:22:17 <licquia> bzr can do that; i believe it's something like "bzr merge -r x..y"
14:22:36 <mwichmann> maybe I'll try
14:23:05 * licquia is ok with it either way; not opposed to backporting, but won't be sad if it happens
14:23:13 <mwichmann> newly created passwd comes back either !! or blank, depending on system config
14:23:15 <licquia> the devel fix is probably the important part
14:23:32 <mwichmann> we're comparing "changed" passwd with original
14:23:41 * licquia also wonders if this isn't an opportunity to prune dead code, but not going to worry about it now
14:23:43 <mwichmann> except we're not
14:24:05 <mwichmann> ... so if password did not change, I think this would incorrectly detect that it changed
14:24:47 <licquia> ah, i see; potential false success
14:25:05 <mwichmann> I believe so
14:25:24 <mwichmann> have to talk these damn things through twice or more before I start seeing the nuances
14:25:30 <mwichmann> just like those network tests
14:25:33 <licquia> in that case, might be better to see if the devel version passes
14:26:07 <licquia> if this has been ignoring a problem all this time, don't want to pile a huge burden on right away
14:26:09 <mwichmann> well, we'd need to jig it up so that "passwd()" fails so we can see if it's detected
14:26:33 <licquia> for now, i'd be happy to just learn that the distros still pass
14:27:13 <mwichmann> let me check my journal full of fails... (throwaway comment from yesterday: had lots of problems with loopback device)
14:27:30 <licquia> target 5.0 w/ 3980
14:27:31 <licquia> ?
14:28:11 <mwichmann> 520|1254 40 745 1 1|passwd - change a user password.  - Reference the Commands and Utilities chapter in the Specification
14:28:11 <mwichmann> 520|1254 40 745 1 2|Password for the user should be changed.
14:28:11 <mwichmann> 520|1254 40 745 1 3|passwd : PASS
14:28:11 <mwichmann> 220|1254 40 0 16:05:51|PASS
14:28:14 <mwichmann> no problem
14:28:35 <mwichmann> yes, that's right for the devel one
14:28:48 <licquia> excellent; i'm thinking we just attach to 5.0, resolve, and call it done
14:29:05 <mwichmann> if you like
14:29:24 <licquia> #agreed resolve 3980, attach to 5.0 release for housekeeping
14:29:32 <mwichmann> seems like the yocto people are using devel tests instead of official 4.1 tests anyway
14:29:46 <mwichmann> not sure of that, but think so
14:33:09 <licquia> ok, that's it for lsb bugs
14:33:14 <licquia> move on to fhs?
14:33:20 <mwichmann> if you like
14:33:25 <mwichmann> I'm so out of touch with fhs...
14:33:31 <licquia> heh
14:33:39 <licquia> mostly, there's a lot of old stuff that needs attention
14:33:43 <mwichmann> right
14:34:33 <licquia> !fhsbug 103
14:34:36 <lsbbot> licquia: 04Bug http://bugs.linuxfoundation.org/show_bug.cgi?id=103 normal, P2, ---, licquia, NEW , Please clarify "removable media"
14:35:52 * licquia reads this over; looks like a philosophy discussion
14:36:03 <licquia> "what does 'removable' really mean?"
14:36:40 * licquia looks at the proposed patch
14:37:11 <mwichmann> it's not bad, not sure it's really necessary....
14:37:24 * licquia is inclined to agree
14:37:40 <licquia> "removable media" really isn't that hard to understand, imho
14:37:45 <mwichmann> no
14:38:16 <mwichmann> if we're to give examples - as the original does, we could be slightly more modern and not mention floppies
14:38:36 <mwichmann> perhaps replace by a flash storage mention
14:38:47 <mwichmann> if we even think it needs changing...
14:40:29 <mwichmann> thoughts?
14:40:33 <licquia> yeah, we could do that, but don't think it's really that unclear
14:41:12 <licquia> #agreed close fhs bug 103 as invalid
14:41:19 <mwichmann> maybe add the word temporary and close?
14:41:25 <mwichmann> okay, no objection
14:41:38 <licquia> !fhsbug 508
14:41:40 <lsbbot> licquia: 04Bug http://bugs.linuxfoundation.org/show_bug.cgi?id=508 enhancement, P2, ---, licquia, NEW , The standard is unclear on what's part of the root-filesystem and what not
14:43:00 <mwichmann> is there anything we need to say wrt the root/usr split?
14:43:04 <licquia> no comments since last clarification
14:43:24 <mwichmann> since of course some trends have been to eliminate that
14:43:30 <licquia> i note that apparently opensuse has followed fedora in making mounted /usr impossible
14:43:57 <licquia> debian-land still explicitly endorses that goal, and ubuntu has not done anything to break it
14:45:21 <mwichmann> seems like fhs can no longer endorse it as a goal, if it actually does
14:45:41 * licquia does a quick search for a bug on the /usr mount goal
14:46:28 <licquia> !fhsbug 940
14:46:31 <lsbbot> licquia: 04Bug http://bugs.linuxfoundation.org/show_bug.cgi?id=940 normal, P2, ---, licquia, NEW , sharing /usr is not a best practice
14:46:40 <licquia> so i think we have that bit covered
14:47:14 * licquia votes to close 508, and worry about /usr in 940
14:48:29 <mwichmann> I guess I don't see anything else here
14:48:46 <licquia> #agreed close fhs 508 as invalid
14:48:59 <mwichmann> e.g. /tmp could be on root, or a tmpfs is more likely now
14:49:05 <mwichmann> too hard to say anything more definitive
14:49:07 <licquia> yup
14:49:36 * licquia moves on to 940, since it came up and is interesting
14:50:59 <mwichmann> so...
14:51:46 <licquia> specifically, that bug is more about sharing /usr across machines, like via nfs
14:51:46 <mwichmann> this is maybe a slightly different issue, but it does poke a hole in the idea of being able to nfs-mount a /usr
14:51:56 <mwichmann> as we say the same thing
14:52:15 <licquia> but i think the problem is more fundamental, in that we can't even move /usr off /
14:52:40 <licquia> (so, one could use a shared root, but not a separate shared /usr)
14:52:52 * licquia writes a "fhs 3.1 rollup" bug
14:55:08 <licquia> now that that's done, i think 940 should probably be on our "next" list
14:55:26 <mwichmann> sounds good
14:57:51 <licquia> #agreed attach 940 to "next" fhs
14:58:42 <mwichmann> ye gods... dependency hell... (I'll leave it in other channel later)
14:59:07 <licquia> !fhsbug 539
14:59:09 <lsbbot> licquia: 04Bug http://bugs.linuxfoundation.org/show_bug.cgi?id=539 enhancement, P3, ---, licquia, NEW , site-local directories (shared across multiple systems)
14:59:54 <licquia> another "requested comments, none came in" bug
15:00:24 * licquia recommends closing
15:01:30 <mwichmann> can always file a new bug
15:01:34 <licquia> yup
15:01:46 <licquia> #agreed close fhs 539
15:02:23 <licquia> ok, think that should do it for today
15:02:28 <licquia> anything else?
15:03:33 <mwichmann> no... I've been distracted anyway (sorry for that) and need to get back to it
15:05:10 <licquia> ok
15:05:13 <licquia> #endmeeting